The production of oxygen using vacuum swing adsorption (VSA) is well-known to air separation technologists. VSA offers a simple non-cryogenic method to produce gaseous oxygen at purities of 80% to 95%. In the last 20 years oxygen VSA plants have become widespread and are offered in various bed configurations. The multi-bed VSA is typically used in the size rage of 60 tons per day (TPD) and higher. The single bed process was adopted as a lower capital, simpler process for lower production ranges, typically 1 TPD up to 40 TPD. Typical single bed systems usually consist of a single blower train that is used for both the feed air provider as well as the regeneration vacuum system. The process usually incorporates automatic valves to direct the air and vacuum flows during the cycle. A newer embodiment of the single bed process uses a reversing blower to generate the feed stream and apply vacuum for the regeneration step. This latest embodiment is well suited for small to medium sized oxygen VSA production plants (1 to 10 TPD). One example of a single bed reversing blower (SBRB) VSA process of this type is described in U.S. Pat. No. 8,496,738.
Although the single bed reversing blower (SBRB) VSA process is simple in practice, its simplicity comes with performance trade-offs when compared to multi-bed systems. Firstly, the lack of additional adsorber beds does not allow for a crucial bed to bed equalization. The pressure equalization step is key to lowering power consumption and increasing product oxygen recovery. Technologists in the art have overcome this deficiency by adding an equalization tank to the SBRB system (such as equalization tanks in SBRB systems provided by Air Liquide of Houston, Tex.).
Typical VSA (and PSA) adsorption systems using molecular sieve for oxygen production are limited to oxygen purity below about ninety-six percent. This is because air is one percent argon and typical adsorption materials neither adsorb oxygen nor argon. Thus, argon is collected with the oxygen and ends up limiting the purity of the oxygen. For many purposes, having generally inert argon within the oxygen is not a particular problem. However, in some instances it is desirable to have oxygen which has a higher purity than ninety-six percent.